The objective of Project 1, Experimental Therapy, is to develop guidelines at the preclinical level for optimization of anticancer drug combination based on principles of biochemical modulation for application in clinic. Modulating agents and effector agents will be combined so that the existing quantitative differences among the metabolic determinant of drug action in tumor cells versus normal host cells will be magnified to favor the antitumor action of the effector agent. Therefore, one modulating agent may be selected for combination with an effector, or cytotoxic, agent in order to increase cytotoxicity specifically in tumor cells as opposed to normal cells, and another modulating agent (e.g., normal metabolite) may be selected for ability to protect normal cells specifically from the cytotoxic action of an effector agent (e.g., an antimetabolite). Agents will be selected on the basis of a specific biochemical rationale and drug combinations will be assembled in a stepwise approach. If an increased antitumor effect is accompanied by untoward host toxicity, the next step is the addition of an agent to selectively protect the host. In addition to the specific "rescue" approach for antimetabolite toxicity with the corresponding normal metabolite, attempts will be made to stimulate more rapid recovery with hematopoietic cytokines, and selected agents will be evaluated to stimulate recovery of intestinal epithelium after drug treatment. This procedure continues with the addition of another drug to yield further augmentation of tumor toxicity, and so on, until the ultimate objective of cure is attained. Specific drug combinations are proposed, including a conceptually new type of modulation (achieved with a 3-drug combination, PALA + MMPR + 6-AN) aimed at amplification of the biochemical changes associated with "programmed cell death' or apoptosis resulting from treatment with many different useful chemotherapeutic drugs. This approach seeks the control of serious host toxicity as essential to the achievement of chemotherapeutic cure, because the resulting operational increase in drug selectivity will allow both a quantitative and a qualitative increase in the chemotherapeutic drug combination. Therapy studies will be performed entirely in vivo murine tumor models. Tumors will be advanced at initiation of treatment, and therapeutic activity will be assessed in terms of tumor growth inhibition, number of partial and complete tumor regressions, and ultimately, the most promising drug combinations will be evaluatd for activity against metastatic tumor and for effect on lifespan. Potential toxic side effects (e.g., damage to the intestinal epithelium and leukopenia) will be monitored in the same experiments. Emphasis in all of the proposed studies is on optimization for therapeutic selectivity, and not merely for potency.